Artificial Consciousness, is the fear rational?
When talking about artificial intelligence, some people are afraid that we may someday create something that will seize power and enslave us. Is this a rational fear?
Some time ago, in response to this fear that Elon Musk seems to be spreading, I wrote a very short article on my findings in my research in Artificial Conciousness, one step beyond Artificial Intelligence.
Today I found this TED talks video of someone who approached the problem from an entirely different angle as I did but came to similar conclusions.
In this steemit post I want to present both; my article:
Artificial Consciousness
Artificial intelligence (AI) today has very little to do with intelligence. These are number crunching machines that evaluate every possible option and choose the best. But that is not at all how our brain works, nor does it even remotely resemble our “thinking”, nor is there any consciousness. For us to produce real AI, it must be able to think freely and in order to do that there must be some awareness, and in order to achieve that it must be self-aware. For this it must have senses through which it can experience itself in relation to its environment. But an AI program is software and software is a virtual thing, it does not exist in our world and it exists only by virtue of being executed (run). We can not hold a program in our hands. Yet to the AI program it must appear in some material form.
When this AI program can see itself it can also see other software components running on the same hardware, it can see in cyberspace, a universe that it has created as a necessity for its own consciousness. Space as our program sees it has no relation to space as we see it.
We can connect a camera to its hardware and this will produce signals that can create changes that it can perceive. But it can not make any sense of it because it does not relate to its universe, nor to itself.
Thinking about this, we could be the result of an artificial consciousness running on some hardware in another universe. Is that what the qabalah refers to as negative existence (ain – ain soph – ain soph aur), this other universe, the hardware that produces us?
And the TED-talks video:
So I think the fear is entirely irrational, yet, thinking about this feels like pulling the supports of our own reality….
What do you think?
Elon Musk and Stephen Hawking both seem to share a fear of AI that's ... very rooted in the eighties. It is unlikely that we'll ever get AIs of the kind they are thinking of. More likely, we'll just have more and more bots running more and more real-world internet-of-things processes and intersecting with each other in ways that we will eventually realize have become wholly incalculable to us. Smart cities and cameras and gait recognition algorithms that can perhaps err and literally "forget" a person, rendering them a non-person in effect.
The end result wouldn't be Skynet or IAM or even Chappie, it'd be more like ten thousand tiny iterations of Lem's Solaris, an alien mirror of our forgotten choices.
A very interesting post @mage00000 , but in my homble opinion.
AI will ended with are humanity , but not with the humanity. Look what humanity have done since we have rational actis . Think about what would happend if we create a human that is in any aspect better than you.
I dont think that this type of technoogy innovation is benefeciary of human race.
Instead my opinion is that whe are going to end all of us, as cybrogs. XD
Thank you for sharing your opinion venela! I guess I am somewhat more optimistic. ;)
Yes thats true , but the only thing that we know is that we dont know the future , so lets life now.
This is a captivating subject. I don't think the fear of AI is totally irrational as they could be many ways its power becomes unmanageable by humans.
"Wait but Why" made a very long article which is my main reference on the subject. Highly recommended reading: https://waitbutwhy.com/2015/01/artificial-intelligence-revolution-1.html
Thanks for the link!
Yes, I guess I should have been more specific. There are also rational fears connected to this subject, but computers are our tools and I don't think that there will come a day that we will become the tools of our computers, which is what I hear in Elon's fear-mongering.
We might not become tools of our computers, but they might have a goal to works towards that is hindered by your existence. If we're not careful and give a super intelligent AI a certain purpose to fulfill,
I think it is illogical to build a machine and tell it to solve a problem like for instance global warming. That is very close to equipping a drone with a gun and program it to kill all humans.
It makes more sense to tell it to find possible solutions, so that we can review those and choose one to implement.
I guess you do have a point, but one must be quite insane to build such a machine....
(yes, I know, more insane people have walked the earth)
Great post a very interesting concept that makes a lot of sense. Kind of warming that robots won’t take us over phew
We all know what happened to Microsoft's chat bot Tay. Considering that we don't need all purpose AI to start potential conflicts. Even simple chat bot on one or both ends could start a nuclear war.
That is especially true in today's world where very few interactions are done face to face. One does simply not know with who he's talking to.
Artificial Intelligence, we think of it in our dreams in old days but now its like dream come true.
The only thing I fear is Elon Musk's tendril like reach. I personally think there are too many variables out there to be controlled (at least at current universal-natural and man-made levels) and that if one is going to be afraid of what you speak of there... there are many equally threatening (non-threatening) things to be afraid of. I hide in the shadows too much as it is ;) The technology is not the thing to fear. The people are. I have faith in humanity but what we are doing with AI and other tech... it's kind of sad.
This goes a bit on the philosophical side of things, but AI can still have execute functions and be AI, because that is how our brains also work. Millions of execute commands are fired in our brain, which together make what we feel as a free will. I'm not going to go into the never ending discussion whether free will exists or not ( I believe it does), but my point is that enough of certain stimuli and interpretation results in choices. Therefore, there exists some AI that actually interpret for themselves based on what happens in the world. They even learn to work with things new to them, things that weren't programmed into them. Of course the learning mechanism needs to be programmed into the AI, but we humans also have innate functions that fire a certain way, if we decide not to act on them, we probably have previous experience that makes us do things otherwise, since we learned from them.
I think at the point we find out if it is bad for us to develop AIs is when they think enough for their selfs. Hopefully, anyone who programs them, will program them to want to help through ethical decisions, not true calculated actions. Anyway, would be stupid not to take the chance.
Thank you for your contribution. There are these so called learning algorithms that respond to new situations in a certain way that can be stimulated by previous experiences. Neural networks can perform that kind of behaviour. But these are still unconscious machines, will-less so to speak. They will remain tools.
I'm afraid of artificial intelligence. With his still not figured out.
You are not the only one, but I think this is mostly fear for the unknown, fear for what you do not fully understand.
I have been working with computers for over 30 years and I have absolutely no fear of them. Yet, it is of course possible to make a drone, give it a gun and program it to kill every human it sees. But even in that case, it remains a tool. A tool for a very sick man... and there are plenty of those. :(